Researchers at NIST have developed a silicon chip that uses light signals to create a miniature neural network grid, potentially revolutionizing artificial neural network design. The chip vertically stacks two layers of photonic waveguides to enable complex signal routing, mimicking neural systems. This 3D design allows for precise control of light intensity distribution between neurons, with applications in quantum information science and autonomous vehicle control systems. The breakthrough demonstrates the feasibility of using light instead of electricity for neural network signaling, which could lead to faster, more efficient processing and enable the development of highly intuitive control systems for autonomous vehicles.
Source: https://www.nist.gov/news-events/news/2018/07/nist-chip-lights-optical-neural-network-demo
Keywords: Neural Networks, Photonic Waveguides, Silicon Chip, Quantum Information Science, Optoelectronic Systems
